GPX3 IS REQUIRED TO MEDIATE PROTECTIVE EFFECTS OF GPX1 LOSS IN COLITIS

Abstract

Abstract Reactive oxygen species (ROS) have long been known to contribute to the development and severity of Inflammatory Bowel Disease. However, more recent data indicates ROS may not always be pathogenic, and in fact ROS may be necessary for intestinal homeostasis. Thus, the role of ROS in the intestine, and antioxidant proteins which regulate ROS, may be contextual and dose-dependent. Members of the glutathione peroxidase (GPx) family are antioxidants that reduce hydrogen peroxide to water. Previously, we determined that loss of GPx1, the most widely expressed GPx, protected against dextran sodium sulfate (DSS)-induced colitis and augmented intestinal proliferation. To better understand how a stepwise increase in ROS may affect intestinal phenotypes, Gpx1-/- mice were next crossed with Gpx3 null mice to yield double-knockout (DKO) mice. First, we assessed baseline proliferation, apoptosis, and DNA damage in the colon. Interestingly, while Gpx1-/- mice displayed increased proliferation compared to WT mice, this was not observed in the DKO mice. Furthermore, Gpx1-/- mice had no change in apoptosis or DNA damage, yet both were increased in DKO mice. Together, these data suggest DKO mice lose the survival benefits of GPx1 loss upon greater reduction of antioxidant capabilities. Next, we assessed the effects of combined GPx1 and GPx3 loss in the context of inflammation and colitis. At baseline, multiplex cytokine profiling identified higher levels of colitis-associated cytokines, including IL-12, CCL4, and CCL5, in DKO mice as compared to either Gpx1-/- or WT cohorts. Pro-inflammatory changes were more pronounced upon aging, as one-year-old DKO mice showed higher levels of TNF-α and IL-17, as well as histologic evidence of chronic inflammation which was absent in similarly aged WT mice. Next, we tested the effects of experimental DSS colitis. Here, while Gpx1-/- mice had less severe injury in the DSS colitis model, DKO mice showed more pronounced weight loss and a trend towards shortened colons as compared to WT mice. Finally, altered ROS-mediated signaling is known to impact interactions with the microbiota, and increased microbial diversity is associated with protection from colitis. To establish whether loss of GPx1 and GPx3 impacted the gut microbiome, stool was harvested from WT, Gpx1-/-, and DKO mice and bacterial DNA was assessed by 16S RNA sequencing. Interestingly, and in line with less severe colitis, Gpx1-/- mice had increased diversity and evenness compared to WT mice, whereas there was no change between WT and DKO mice. Taken together, these data indicate that GPx3 is required to mediate the anti-inflammatory effects of GPx1 loss. Furthermore, combinatorial GPx loss induces greater colonic inflammation at baseline and during DSS-induced colitis, highlighting dose-dependent effects of ROS and antioxidant capacity in the intestine.